Download Mortality Study of the Significance of Extrasystoles in an

Mortality Study of the Significance
of Extrasystoles in an Insured Population
By MANUEL RODSTEIN, M.D., LEON WOLLOCH, M.A.,
AND
RICHARD S. CUBNER, M.D.
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
SUMMARY
The prognostic significance of extrasystoles was evaluated in 712 insured persons
with this finding who were followed for an average period of 18 years. At the end of
the study 356 persons were alive and 125 were dead. Two hundred and thirty-one
persons had terminated their policies at a prior date by maturity or discontinuance
and were not followed further. Although the lack of follow-up to 1968 of this latter
group imposes limitations on the results and conclusions of the study, their exclusion
resulted in no significant change in the mortality ratio of the remaining group of 481
persons followed to death or the anniversary of their policies in 1968.
The observed mortality ratios showed no appreciable increase in a group of 604
individuals who were considered to have a normal life expectancy except for the
presence of extrasystoles at the time of evaluation for insurance. Similar normal mortality ratios were observed in subgroups with simple and complex supraventricular
and ventricular extrasystoles. There was no appreciable difference in mortality when
those under 40 and those 40 years of age and over at the time of discovery of extrasystoles were compared. There was no significant difference in mortality between
those who responded to exercise with an increase in the number of extrasystoles and
those who had no change or a decrease immediately after exercise. An increase in the
number of extrasystoles per minute was associated with an increase in mortality.
There was a low incidence of sudden death among those with supraventricular
extrasystoles.
Ventricular extrasystoles, particularly those of complex types, were associated with
a high incidence of death from myocardial infarction, often of a sudden nature, particularly in persons under the age of 56 years. There were too few cases of the
individual types of complex ventricular extrasystoles to determine whether this was
a characteristic of the group as a whole or of one or more of its constituent types.
Ventricular extrasystoles in the presence of other cardiac abnormalities andlor elevation of the blood pressure were associated with an increase in mortality. The mortality experience in such persons was distinctly higher than in those with similar
degrees of substandard expected mortality whose abnormalities were not of a cardiovascular nature.
Additional Indexing Words:
Simple extrasystole
Mortality ratio
Substandard risk
Complex extrasystole
Standard risk
Myocardial infarction
Exposure years
T wave
death. This, together with recognition of the
great frequency of ectopic ventricular beats in
monitored patients with acute myocardial
C ONSIDERABLE interest has attended a
number of recent reports that follow-up
studies of persons with ventricular premature
beats show a heightened incidence of sudden_____________________
Address for reprints: Manuel Rodstein, M.D., The
Equitable Life Assurance Society of the United States,
1285 Avenue of the Americas, New York, New York
10019.
Received March 29, 1971; revision accepted for
publication June 7, 1971.
From the Medical Department of the Equitable
Life Assurance Society of the United States, 1285
Avenue of the Americas, New York, New York
10019.
Circulation, Volume XLIV, October 1971
617
RODSTEIN ET AL.
618
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
infarction and their significance as harbingers
of the more serious arrhythmias of ventricular
tachycardia and fibrillation, has raised further
question whether ventricular premature beats
are as benign as has been generally considered.
The study herewith reported extends an
investigation of the significance of extrasystoles among an insured population reported in
1941, updating and extending this material,
which at that time did not allow conclusion as
to mortality, although 1,142 persons with
extrasystoles were studied, because of the
paucity of deaths.' In brief, it was found then,
and the findings are confirmed in this study,
that ventricular extrasystoles are much more
frequent than supraventricular extrasystoles.
The present study documents that mortality
experience in persons with atrial and other
supraventricular extrasystoles is normal. What
has not been previously investigated, and is
shown by the data of the present study, is that
mortality experience of persons with ventricular extrasystoles as an isolated finding likewise
is relatively normal.
Methods
This study is based on 712 individuals insured
by the Equitable Life Assurance Society of the
United States between 1930 and 1956 in whom
one or more extrasystoles were found in the
electrocardiogram. Six hundred and four persons
were considered standard risks for insurance, that
is, having a normal life expectancy if extrasystoles
had been absent. One hundred and eight persons
were considered substandard risks for insurance,
that is, having a decrease in life expectancy of up
to approximately 7 years due to other abnormalities even if the extrasystoles had been absent.
The persons were selected from approximately
30,000 individual applicants for insurance between 1930 and 1956 in whom one or more
electrocardiograms had been reviewed in the
Home Office Medical Department.
The following data were recorded: age at time
of discovery of extrasystoles, age at time of
examination for insurance, pulse rate, number of
extrasystoles per minute, abnormalities other than
extrasystoles, cause of death, and duration of the
terminal illness as recorded on the death
certificate. In the great majority of persons, the
electrocardiogram with extrasystoles was recorded
on the day of examination for insurance or within
a short interval of time thereafter.
The persons in the study were observed from
the year the electrocardiogram was recorded to
the anniversary of this date in 1968 if they were
alive or the year of termination of the policy by
death, maturity, or discontinuance prior to this
date. Thirty-one of the persons were declined for
insurance; they were followed in an identical
manner as the other 681 persons through policies
previously issued to them by the Society. No
attempt was made to follow up persons whose
policies were terminated for reasons other than
death. At the end of the study 356 persons were
alive, 125 were dead, and 231 had terminated
their insurance policies at a prior date by maturity
or discontinuance. The duration of observation
ranged from a minimum of 1 year to 45 years,
resulting in 12,900 exposure years, an average of
18 years per person.
The group was made
up almost completely of
the time of admission to
was from 17 to 65 years. Twenty-five
persons were under 24 years, 178 were 25 to 35
years, 258 were 35 to 45 years, 190 were 45 to 55
years, and 61 were 55 through 65 years of age at
this time.
Actuarial principles were applied in the
computation of the expected number of deaths by
means of the Male Intercompany 1955-1960
Basic Select Tables and the Male Intercompany
1957-1960 Basic Ultimate Tables. These tables
were compiled from data supplied by the 16
largest insurance companies in the United States.*
The mortality of the insurance applicants is
The
the study
men.
age range at
*In effect, persons insured by the insurance industry
constitute the control groups for the computation of
the expected number of deaths. They consist of: (a)
persons who had been insured at standard rates for
not more than 15 years and whose mortality was
observed between 1955 and 1960 for each age and
duration of exposure (the Male Intercompany
1955-1960 Basic Select Tables); and (b) persons who
had been insured at standard rates at least 15-years
prior to their entry into the period of observation,
1957 to 1960 (the Male Intercompany 1957-1960
Basic Ultimate Tables). In the second group, any
effects of the initial medical examination in lowering
mortality during the early policy years are no longer
present and the rate of mortality becomes a function
of the person's attained age. Thus, expected deaths
were calculated for cohorts similar in age distribution
and length of follow-up to the entrants presented in
the various tables.
As a substantial portion of the exposure in this
series occurred between 1930 and 1955, when
mortality rates were higher than in the 1955 to 1960
base period used in this study for calculating the
number of expected deaths, the mortality rates we
have presented are presumably overstated.
Circulation, Volume XLIV, October 1971
MORTALITY STUDY
619
measured by dividing the number of actual
close to the preceding T wave (within 0.04
second of the termination of the T wave); and
bigeminy or trigeminy. Unifocal extrasystoles
without any of these characteristics were classified as simple.
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
deaths by the number of expected deaths. The
resulting quotient multiplied by 100 is a
percentage known as the mortality ratio and is the
measuring rod used in all actuarial comparisons of
mortality. A group in which the number of actual
deaths equals the number of expected deaths has
a mortality ratio of 100%. Persons insured at
standard rates are used as a control group and
their mortality is taken as 100%. Therefore, for
example, a ratio of actual to expected deaths of
150% of a subgroup would indicate that the
mortality of those in question is 50%; greater than
that in the control group of the same age
distribution over the period of observation.
Extrasystoles were classified as supraventricular, including those of atrial and atrioventricular
nodal origin, and as ventricular. They were
further subdivided into categories of simple and
complex. Complex extrasystoles were those with
the following characteristics: aberrant conduction
(supraventricular only); paired, multifocal origin,
with postextrasystolic T-wave inversion on or
Results
The mortality ratio of the 604 persons
vithout other significant abnormalities was
102% (table 1). The difference in mortality
ratios between the subgroups of simple and
complex extrasystoles was not of statistical
significance. Comparison of those under 40 and
those 40 years of age and over revealed no
significant difference in mortality ratios.
Subgroups of persons with various types of
complex extrasystoles without other abnormalities were too small for determination of
meaningful mortality ratios, with the exception of the group in whom the ventricular
Table 1
Mortality of Individuals with Extrasystoles without Other Abnormalities*
Type of extrasystole
Supraventricular
Simple
Complex
Ventricular
Simple
Complex
All
Entrants
Actual
deaths
Expected
deaths
Ratio actual to
expected deaths
1a
12.69
6.22
118%
96%o
59.79
17.34
96.04
121 %
83
39
6
56
359
123
604
21
98
94%O
102%'7
(836c
to
123%)t
*Standard insurance risks in the absence of extrasystoles.
t95% confidence limit.
Table 2
AMortality of Individuals with Extrasystoles without Other Abnorrnalities* by Type of
Ventricular Extrasystole
Actual
deaths
Expected
deaths
expected deaths
56
5
59.79
6.68
94%
75%
59
4
12
138%
482
77
1.99
8.68
77.14
Type of extrasystole
Entrants
Simple
Bigeminy or trigeminy
Paired, multifocal, and with
postextrasystolic T inversion
On or close to preceding T wave
359
50
14
*Standard insurance risks in the absence of extrasystoles.
tA/E ratios not shown for fewer than five deaths.
t9.5%/ confidence limit.
Circulation, Volume XLIV, October 1971
Ratio actual to
t
100%
(786,7o to 122(7o)t
RODSTEIN ET AL.
620
When deaths from accident and suicide
excluded, 38% of the entire group died
from acute myocardial infarction. Thirteen of
those with complex ventricular extrasystoles
(54%), 21 of those with simple ventricular
extrasystoles (34%o), and four of those with
supraventricular extrasystoles (27%) died of
acute myocardial infarction (table 4). At the
time of death from acute myocardial infarction the average age of the group with complex ventricular extrasystoles was 55.8 years;
seven of the 13 persons were under 56 years
of age. The average age at the time of death
from acute myocardial infarction of the group
with simple ventricular extrasystoles was 61
years; 10 of 21 persons were under 56 years of
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
extrasystoles were on or close to the T wave of
the preceding normal beat. There were 59
such persons, among whom there were 12
deaths and the mortality ratio was 138%
(table 2).
The mortality ratio of the 108 persons with
extrasystoles and other abnormalities was 167%
(table 3A). Among such associated abnormalities were blood pressure elevation, localized
apical systolic murmurs, overweight, elevation
of blood sugar, hernias, enteritis, gastritis,
colitis, migraine, sciatica, spinal disc disease,
and varicose veins. The subgroup with ventricular extrasystoles together with blood
pressure elevation and/or a cardiac abnormality had a mortality ratio of 223%. The mortality
ratio of the subgroup with ventricular extrasystoles together with noneardiovascular abnormalities was 131%. The difference between
these two mortality ratios, while suggestive, is
not statistically significant (table 3B).
were
age.
Among the 21 persons who had ventricular
extrasystoles with fatal acute myocardial
infarction in whom the duration of the
terminal event was noted, sudden death
Table 3A
Mortality of Insured Individuals with Extrasystoles and Other Abnormalities*
Entrants
Actual
deaths
Expected
deaths
Ratio actual to
expected deaths
14
8
2
2
1.79
1.54
t
t
62
24
108
15
8.71
4.08
16.12
172%C
196%
Type of
extrasystole
Supraventricular
Simple
Complex
Ventricular
Simple
Complex
All
8
27
167%n
(110% to 242%)t
*Rated up to 195%c expected mortality for conditions other than extrasystoles.
tA/E ratios not shown for fewer than five deaths.
t95%/C/ confidence limit.
Table 3B
Mortality of Individuals with Ventricular Extrasystoles and Other Abnormalities*
Type of other abnormality
Rating for blood pressure
and/or cardiac condition
No rating for blood pressure
or cardiac condition
All
Entrants
Actual
deaths
42
15
6.70
223%
44
86
8
23
6.10
12.80
131%
180%
Expected
deaths
Ratio actual to
expected deaths
(114% to 269%)t
*Rated up to 195% expected mortality for conditions other than extrasystoles.
t95% confidence limit.
Circulation, Volume XLIV, October 1971
621
MORTALITY STUDY
Table 4
Relation of Types of Antecedent Extrasystoles to
Cause of Death (Death Due to Acute Myocardial
Infarction)
Type of extrasystole
Number
Supraventricular
Ventricular
Simple
Complex
All
4
Percent of all causes*
without other abnormalities. It consisted of 15 vigorous hops or stepping on and
off an ordinary chair 15 times in 1 min, with
the pulse rate reported before, immediately
after, and 2 min after exercise.
persons
Discussion
27
21
13
38
34
54
38
*Excluding deaths due to accident, suicide, and in
which the cause of death was not ascertained.
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
occurred in 16. Fourteen of 15 persons with
ventricular extrasystoles who died of acute
myocardial infarction under the age of 56
years had neither elevation of blood pressure
nor other evidence of cardiac disease at the
time of examination for life insurance. Of the
22 sudden deaths in the entire series, only two
occurred among those with supraventricular
extrasystoles.
An increase in the frequency of extrasystoles
from less than 10 to 10-30 per minute was
accompanied by an increase in mortality from
92% to 123%. Among those aged 45 years and
over this increase was from 104'% to 148%
(table 5).
The persons with an increase in the number
of extrasystoles immediately after exercise,
and those with a decrease or no change, had
similar mortality ratios, 96% and 105%, respectively (table 6). This test was performed in 333
Following accepted actuarial practice, the
231 persons who terminated their insurance
policies at a date prior to 1968 by maturity or
discontinuance were included in the study for
as many years as we had information about
them, an average of 12.2 years. This lack of
follow-up imposes certain limitations on the
results and conclusions of the study. If this
group had been excluded from the study the
mortality ratio for the remainder would have
been 126%, rather than the figure of 102% found
for the entire group. This difference is
insignificant at the 95% confidence level.
Among the applicants for insurance who
were considered to have a normal life
expectancy except for the presence of extrasystoles, the observed mortality ratios for the
entire group and for the subgroups with
simple and complex supraventricular and
ventricular extrasystoles were essentially normal, ranging between 94% and 121% and falling
within the standard insurance-premium classification. These findings are consistent with
the opinion of Campbell2 that the presence of
extrasystoles in the absence of other evidence
of heart disease has little prognostic significance. Thus, the group of cases with extrasys-
Table 5
Mortality of Individuals with Extrasystoles without Other Abnormalities by Number
of Extrasystoles
Number of
per
Minute
extrasystoles
Issue age
(years)
Less than 10
Under 45
45 and over
All ages
Under 45
45 and over
249
151
62
All ages
213
604
per minute
10 and over
Total
Entrants
142
391
Actual
deaths
Expected
deaths
25
36
19
18
31.55
34.53
66.08
17.79
12.17
37
98
29.96
96.04
61
Ratio actual to
expected deaths
79%
104%
92%
107%
148%
123%
102%
(81% to 123%)*
*95%
confidence limit.
Circulation, Volume XLIV, October 1971
RODSTEIN ET AL.
622
Table 6
Mortality of Individuals with Extrasystoles without Other Abnormalities by Response
to Exercise
Issue age
(years)
Entrants
Under 43
43 and over
All
Expected
deaths
Actual
deaths
Ratio actual to
expected deaths
Decreased or Remained Same after Exercise
24.40
181
19
21.43
90
29
4.3.83
48
271
78%,
135%7O
105%7^
(73%Ec to 135%)
Under 45
43 and over
All
46
16
62
Increased
5
3
10
7.74
2.71
10.43
65%
185%
(43%,
96%c
180%,)*
to
*93 confidence limit.
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
toles is apparently so heavily weighted with
normal individuals whose extrasystoles are of
physiologic origin that on the whole the
finding does not carry a significant adverse
prognosis. The number of those with undetected or subelinical cardiovascular disease
with pathologic causes for their extrasystoles is
apparently too small to affect the prognosis to
a significant degree.
These results are similar to those of studies
conducted in the Medical Department of the
Equitable Life Assurance Society with regard
to the effect on mortality of electrocardiographic low T waves and right bundle-branch
block in otherwise normal applicants for
insurance.3 4 There are a few comparable
studies; the results are similar. Brandon et al.
found a mortality ratio of 141% among all
applicants for insurance to the Aetna Life
Insurance Company with normal electrocardiograms except for the presence of extrasystoles, irrespective of the number and degree of
other physical impairments.5 Lyle conducted a
mortality study among 11,000 Home Office
employees of the Prudential Life Insurance
Company over the period of 1933 to 1960. No
differentiation was made into groups with
supraventricular and ventricular extrasystoles.
Among males without other abnormalities the
mortality ratio was 115% for those with simple
extrasystoles and 130% for those with complex
extrasystoles.6 Comparable groups in our
series had mortality ratios of 98% and 115%,
respectively.
The discovery of extrasystoles at a later age
was not related to an increase in mortality in
this study. This is contrary to the assumption
that a higher mortality would be expected in
older individuals since in them extrasystoles,
particularly those of ventricular origin, are;
more likely to be manifestations of subelinical
and undetected coronary artery disease. Our
findings may be due to the absence of an
appreciable number of individuals over 54
years of age at the time of discovery of
extrasystoles. The results in Lyle's series were
similar. The increase in mortality noted in this
series with an increased frequency of extrasystoles, particularly among those over 45 years
of age, is similar to the findings in Lyle's
series."
The group of persons in whom there was an
increase in the number of extrasystoles immediately after exercise demonstrated no
significant increase in mortality. The mortality
ratio of 185% among those 45 years of age and
over who had an increase in the number of
extrasystoles with exercise is suggestive. It is
not significant, as it is based on only five
deaths among 16 entrants. Lamb and Hiss
foud that in the absence of definite supporting
evidence of valvular disease, angina, or other
signs of coronary artery disease an increase in
the number of extrasystoles after exercise, or
Circulation, Volume XLIV, October ]971
MORTALITY STUDY
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
their appearance de novo, is not indicative of
significant heart disease.7 Mann8 found no
evidence of coronary insufficiency in the
majority of persons in whom extrasystoles
present at rest persisted or became more
frequent after exercise; he found evidence of
coronary insufficiency in all 21 persons in
whom extrasystoles appeared for the first time
after exercise.
The normal mortality ratios in the entire
group and in the subgroups of simple and
complex supraventricular and ventricular extrasystoles in this study do not negate the fact
that extrasystoles may be manifestations of
acute or chronic coronary artery disease and
myocardial involvement from a variety of
causes. Nor do they contradict the findings of
epidemiologic studies such as those of Chiang
et al.,9' 10 Hinkle et al.,1" and Pell and
D'Alonzo,12 which indicate a relationship
between the finding of extrasystoles, evidence
of coronary disease on examination, and an
increased tendency to death, particularly of a
sudden nature, from acute myocardial infarction.
The number of persons in our group with
supraventricular extrasystoles who died of
acute myocardial infarction or had associated
cardiovascular abnormalities was too small for
meaningful analysis. However, of the 24
persons in the entire series in whom the
duration of the terminal cardiovascular illness
was noted as sudden, only two had prior
supraventricular extrasystoles. This finding is
in accordance with both the Tecumseh
study9"10 and Hinkle's study" in which
sudden death among those with prior supraventricular extrasystoles was rare. In these
studies, supraventricular extrasystoles were
not associated with evidence of coronary
artery disease at the time of initial examination, and no increase in the number of
expected coronary deaths occurred in the
period of follow-up. The considerable preponderance of sudden deaths in those with
preexisting ventricular extrasystoles versus
those with preexisting supraventricular extrasystoles in the present study suggests that
ventricular premature beats signify a predisCirculation, Volume XLIV, October 1971
623
position to electrical instability of the ventricles, so that when a major ischemic episode
occurs fatal arrhythmias may ensue.
A mortality ratio of 223% among those who
had ventricular extrasystoles together with
minor elevation of blood pressure and/or
other cardiac abnormality compared with the
131% mortality ratio among those with ventricular extrasystoles together with noncardiovascular abnormalities suggests that those ventricular extrasystoles associated with cardiovascular
disease have an adverse effect on mortality. In
Lyle's series similar results were found in a
group of persons in whom extrasystoles and
elevation of blood pressure coexisted.
Another adverse effect of ventricular extrasystoles is shown by the death from acute
myocardial infarction of 54% of those with
complex ventricular extrasystoles when deaths
from accident and suicide were excluded.
Approximately half died under age 56. This
contrasts with a figure of 34% for those with
simple ventricular extrasystoles, which is
borderline for a male group of this age
distribution. No cardiovascular abnormality
was found in the great majority of both
groups at the time of initial examination,
including 14 of the 15 persons who subsequently died of acute myocardial infarction
under the age of 56. The implications of these
findings are in accord with the results of Pell
and D'Alonzo, who found that in a 6-year
follow-up study of 86,570 employees of the
DuPont Corporation those with antecedent
ventricular extrasystoles had a significantly
higher early, 30-day, case-fatality ratio from
acute myocardial infarction.'2
In the Tecumseh study there was a
significantly higher incidence of coronary
artery disease among those over 30 years of
age with ventricular extrasystoles, and a
significant increase in the incidence of sudden
death among those with antecedent ventricular extrasystoles.9 The findings from the
Tecumseh study have been criticized because
only one of the sudden deaths occurred
among persons under 70 years of age without
preexistent coronary artery disease. In addition, in the age group 40 to 69 years the
6,24
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
incidence of sudden death in those with prior
ventricular extrasystoles was 2.8% (three of 107
persons) compared with 1.3% (24 of 1,835
persons) without preexistent ventricular extrasystoles. This was not considered to be a
marked difference.13 Hinkle et al., employing
6-hour periods of continuous electrocardiographic recording, also found ventricular
extrasystoles to be associated with increased
evidence of coronary artery disease at the time
of first examination and an increased risk of
subsequent coronary death, often of a sudden
nature."
The effect on mortality of the various types
of complex ventricular extrasystoles is a
subject of great current interest, as they are
infrequent in normal individuals, are frequently associated with organic heart disease, and
are often precursors of sudden death from
ventricular tachyarrhythmias when they occur
in the course of acute myocardial infarction.
Lown and Ruberman have recently emphasized the urgent need for thorough studies of
ventricular extrasystoles as to type, frequency,
multiform appearance, site of origin, repetitive patterns, and relation to sudden death.14
Ventricular extrasystoles frequently trigger
ventricular tachyarrhythmias in the necrotic,
hypoxic, hyperkalemic, and acidotic myocardium of an individual with an acute myocardial infarction, but this does not signify that
they are necessarily as dangerous in an
otherwise normal individual. However, one
might hypothesize that if they were a
recurrent or persistent phenomenon in a given
individual over the course of many years they
would assume a more ominous significance
when an acute heart attack supervened.
Unfortunately, the number of persons with
each type of complex ventricular extrasystoles
in this study is too small to answer these
questions.
Among the 123 persons with complex
ventricular extrasystoles without other abnormalities, there were only 14 persons and four
deaths with extrasystoles which were either
paired, multifocal, or of the type with
postextrasystolic T-wave inversion. This is too
small a group for the calculation of a
RODSTEIN ET AL.
meaningful mortality ratio. Among 67,375
asymptomatic Air Force subjects Hiss et al.
found only three individuals with multifocal
ventricular extrasystoles.'5 Many authors have
commented on the frequent association of this
type of extrasystole with evidence of organic
heart disease.2 1618 Levine et al. noted the
rarity of ventricular extrasystoles with postextrasystolic T-wave inversion in the absence
of organic heart disease,'9 also observed by
Ungerleider and Gubner.'
There were 50 persons and five deaths in the
group with ventricular extrasystoles manifesting as bigeminy or trigeminy. The mortality
ratio was 75%. This figure is not statistically
significant, but is in accord with the statements by Katz and Pick20 and Massie and
Walsh'8 to the effect that such extrasystoles
are frequently seen in normal persons. However, Hinkle et al. found them associated with
a significantly greater number of deaths than
expected." Hiss et al. noted their rarity in
asymptomatic Air Force personnel.'5
Among 59 persons with ventricular extrasystoles occurring on or close to the T wave of
the preceding normal beat there were 12
deaths, a mortality ratio of 138%. This type of
ventricular extrasystole, falling as it does on or
near the vulnerable phase of the T wave or in
the supernormal phase following it, has been
considered extremely dangerous, particularly
during the course of acute myocardial infarction, as these extrasystoles may trigger ventricular fibrillation and sudden death. In addition,
such extrasystoles have been reported as
frequently associated with other evidence of
organic heart disease.2' 27 The modest increase
in mortality associated with such extrasystoles
in our study indicates that they are not as
dangerous in otherwise normal individuals.
Finally, comment may be made on the
objection to this type of study based on the
known variability in occurrence of extrasystoles in routine electrocardiograms and the
finding that of 283 persons monitored for 6hour periods only 22 failed to show at least
one extrasystole, which, however, represents
an average of 30,472 beats per individual."1
These findings do not detract from the
Circculation, Volume XLIV, October 197'1
62S
MORTALITY STUDY
significance of the occurrence of extrasystoles
in a routine electrocardiogram since, as noted
by Hiss, while "few individuals ever live an
entire lifetime without an occasional premature contraction, the probability of a chance
prematurity occurring during the brief duration of a routine electrocardiogram is quite
small."15
Acknowledgment
We wish to thank Dr. Leon Warshaw, Vice
President and Chief Medical Director of the Equitable
Life Assurance Society of the United States, for his
most valuable advice and suggestions.
1.
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
2.
3.
4.
5.
6.
7.
8.
9.
References
UNGERLEIDER HE, GUBNEK RS: Extrasystoles and
the mechanism of palpitation. Trans Amer
Therap Soc 41: 1, 1941
CAMPBELL M: Etiology and significance of
extrasystoles. Brit Heart J 79: 142, 1929
RoDsTEnI M, WoLLocH L, GUBNER RS: Significance of minor electrocardiographic T wave
changes. Amer J Med Sci 252: 21, 1966
RODSTEIN M, GuBNER RS, MILLS JP, LOVELL JF,
UNGERLEIDER HE: Mortality study in bundle
branch block. Arch Intern Med (Chicago) 87:
663, 1951
BRANDON KF, NEILL MH, STREErER GC: Use of
the electrocardiogram in twenty-five years of
insurance selection. Trans Ass Life Insur Med
Dir Amer 34: 153, 1950
LYLE AM: Study of premature beats by
electrocardiogram. Trans Soc Actuaries 14:
493, 1953
LAMB LE, Hiss RG: Influence of exercise on
premature contractions. Amer J Cardiol 10:
209, 1962
MANN RH: Premature ventricular contractions
and exercise. Mayo Clin Proc 27: 383, 1952
CHANG BN, PERLMAN LV, OSTRANDER LD JR,
EPSTEIN FH: Relationship of premature
systoles to coronary heart disease and sudden
death in the Tecumseh epidemiologic study.
Ann Intern Med 70: 1159, 1969
10. CHIANG BN, PERLMAN LV, FULTON M, OSTRANDER LD JR, EPSTEIN FH: Predisposing
factors in sudden cardiac death in Tecumseh,
Michigan. Circulation 41: 31, 1970
1 1. HINKrLE LE JR, CARVER ST, STEvENs M:
Frequency of asymptomatic disturbances of
cardiac rhythm and conduction in middle-aged
men. Amer J Cardiol 24: 629, 1970
Circulation, Volume XLIV, October 1971
12. PELL S, D'ALONZO CA: Immediate mortality and
five year survival of employed men with a first
myocardial infarction. New Eng J Med 270:
915, 1964
13. HBB: Minuscle review [of article cited in
reference 9]. Circulation 40: 828, 1969
14. LowN B, RUBERMAN W: The concept of
precoronary care. Mod Conc Cardiovasc Dis
39: 97, 1970
15. Hiss RG, AVERILL K, LAMB LE: Electrocardiographic findings in 67,375 asymptomatic
subjects: III. Ventricular rhythms. Amer J
Cardiol 6: 96, 1960
16. D'IRSAY S: On the meaning of extrasystoles.
Amer J Med Sci 174: 96, 1927
17. BREALL WS: Premature ventricular contractions:
Clinical implications. J Amer Geriat Soc 14:
639, 1966
18. MASsIE E, WALSH TJ: Clinical Vectorcardiography and Electrocardiography. Chicago, Year
Book Publishers, 1960
19. LEVINE HD, LOWN B, STREEPER RB: Clin-
20.
21.
22.
23.
ical significance of postextrasystolic T-wave
changes. Circulation 6: 538, 1952
KATZ LN, PICK A: Clinical Electrocardiography.
Part I. The Arrhythmias. Philadelphia, Lea &
Febiger, 1956
SMIRK FH, PALMER DG: A myocardial syndrome: With particular reference to the
occurrence of sudden death and of premature
systoles interrupting antecedent T waves. Amer
J Cardiol 6: 620, 1960
READ WO: Ventricular arrhythmias. Minn Med
12: 1284, 1965
DOLARA A: Early premature ventricular beats,
repetitive ventricular response and ventricular
fibrillation. Amer Heart J 74: 332, 1967
24. CASTELLANOs A, LEMBERG L, ARCEBAL AG,
AZUCENA G, BERKOvITs BV, HERNANDEZPinREn 0: Repetitive firing produced by
pacemaker stimuli falling after the T wave.
Amer J Cardiol 25: 247, 1970
25. SOLOFF LA, FEWELL JW: The supernormal phase
of ventricular excitation in man: Its bearing on
the genesis of ventricular premature systoles
and a note on atrial ventricular conduction.
Amer Heart J 59: 869, 1960
26. LEPESCHKIN E: The U wave of the electrocardiogram. Mod Conc Cardiovasc Dis 38: 39,
1969
27. HE mNArEz-PERIrn 0, MORALES-ROCHA J,
BARCELO JE: Supernormal phase of conduction
in human heart demonstrated by subthreshold
pacemakers. Brit Heart J 31: 553, 1969
Mortality Study of the Significance of Extrasystoles in an Insured Population
MANUEL RODSTEIN, LEON WOLLOCH and RICHARD S. GUBNER
Downloaded from http://circ.ahajournals.org/ by guest on June 14, 2017
Circulation. 1971;44:617-625
doi: 10.1161/01.CIR.44.4.617
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX
75231
Copyright © 1971 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is
located on the World Wide Web at:
http://circ.ahajournals.org/content/44/4/617
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles
originally published in Circulation can be obtained via RightsLink, a service of the Copyright
Clearance Center, not the Editorial Office. Once the online version of the published article for
which permission is being requested is located, click Request Permissions in the middle column
of the Web page under Services. Further information about this process is available in the
Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation is online at:
http://circ.ahajournals.org//subscriptions/